otto w. gustafson licensing manager · (weld number 5-675) consisted of an intermediate post weld...

EtAEntergy Entergy Nuclear Operations, Inc.Palisades Nuclear Plant27780 Blue Star Memorial HighwayCovert, MI 49043Tel 269 764 2000

Otto W. GustafsonLicensing Manager

PNP 2014-021

March 1,2014

U. S. Nuclear Regulatory CommissionATTN: Document Control DeskWashington, DC 20555-0001

SUBJECT: Response to Request for Additional Information dated February 26,2014, for Relief Request Number RR 4-18 - Proposed Alternative, Useof Alternate ASME Code Case N-770-1 Baseline Examination

Palisades Nuclear PlantDocket 50-255License No. DPR-20

References: 1. Entergy Nuclear Operations, Inc. letter PNP 2014-015, ReliefRequest Number RR 4-18 - Proposed Alternative, Use of AlternateASME Code Case N-770-1 Baseline Examination, datedFebruary 25, 2014

2. NRC Electronic Mail, Request for Additional Information - Palisades- RR 4-18 - Proposed Alternative, Use of Alternate ASME CodeCase N-770-1 Baseline Examination - MF3508, datedFebruary 26, 2014

Dear Sir or Madam:

In Reference 1, Entergy Nuclear Operations, Inc. (ENO) requested Nuclear RegulatoryCommission (NRC) approval of the Request for Relief for a Proposed Alternative for thePalisades Nuclear Plant (PNP). NRC approval was requested by March 8, 2014.

Reference 1 is associated with the use of an alternative to the requirements of theAmerican Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Code Case N-770-1, as conditioned by 10 CFR 50.55a(g)(6)(ii)(F)(1) and10 CFR 50.55a(g)(6)(ii)(F)(3), dated June 21, 2011.

In Reference 2, the NRC issued a request for additional information. Responses to thequestions are provided in Attachments 1 and 2, and the Enclosure.

This submittal contains no proprietary information.

PNP 2014-021

This submittal makes no new commitments or revisions to previous commitments.

Sincerely,

owg/jse

2Attachments: 1. Response to Request for Additional Information for Relief

Request Number RR 4-18 - Proposed Alternative, Use of AlternateASME Code Case N-770-1 Baseline Examination

2. Dominion Engineering, Inc. Letter No. L-4199-00-01

Enclosure: ENO Response to Question RAI-1.11

cc: Administrator, Region Ill, USNRCProject Manager, Palisades, USNRCResident Inspector, Palisades, USNRC

ATTACHMENT 1

Response to Request for Additional Information for Relief Request NumberRR 4-18 - Proposed Alternative, Use of Alternate ASME Code Case N-770-1

Baseline Examination

By letter dated February 25, 2014, Entergy Nuclear Operations (ENO) requestedNuclear Regulatory Commission (NRC) approval of the Request for Relief for aProposed Alternative for the Palisades Nuclear Plant (PNP). By electronic mail, datedFebruary 26, 2014, the Nuclear Regulatory Commission (NRC) requested additionalinformation. The requested information is provided below.

1. NRC Information Request - Response to Question RAI-1.1

How was the heat treatment performed in the field?

ENO Response

All welds referenced in this relief request were post weld heat treated at thevendor's facility, during original fabrication. Per Combustion Engineering detailweld procedure (MA-41, Revision 0), heat treatment of the hot leg nozzle weld(weld number 5-675) consisted of an intermediate post weld heat treatment at11 00°F, - 0°F/+ 500F, for 15 minutes, followed by 11 50'F, +/- 25°F, for one hourper inch thickness of weld. No heat treatment was performed at the site for thesubject welds.


How was the heat treatment modeled?

ENO Response

The heat treatment was modeled by linearly ramping the surface temperature ofthe finite element analysis up and down between 70°F and 1125°F at a rate of200°F/hour. The holding time between the heatup and cooldown ramps is fourhours.

During heat treatment, creep strain relaxation modeling was included when themodel temperature was above 800 0F. The creep model was based on a powerlaw:

dt-- = Au'"dt

1 of 15

Material Temperature A(OF) (ksilhr) n

800 1.26E-13 5.40SA-516 Gr. 70

900 3.59E-14 7.80

(Based on 1000 2.43E-12 10.321/2Mo steel)

1100 2.50E-07 4.11

800 7.73E-19 7.95ER308L

900 5.67E-17 7.42

(Based on 1000 1.82E-13 5.41Type 304)

1100 8.62E-12 4.78

Alloy 600 800 1.50E-19 8.00

Alloy 82/182 900 2.87E-14 5.21

(Based on 1000 3.02E-10 3.21Alloy 600) 1100 1.72E-09 3.32


What was the basis for the 5-cycle shakedown - typically it is not included inweld residual stress calculations?

a.b.C.

Why was 5 cycles used versus two or three cycles for example?How was each shakedown modeled?What is the effect of the shakedowns as they were applied? (i.e. provideweld residual stresses for each application of the technique.)

ENO Response

Operational cycles are frequently included in welding residual stress calculationsas a part of determining the operating stress condition. In particular, thestandard modeling practice adopted by the xLPR (Extremely Low Probability ofRupture) welding residual stress team, which includes the NRC, nationallaboratory, and industry participants, specifies that the welded configurationshould be cycled between operating conditions and residual conditions to shakedown the nonlinear material hardening behavior. Typically, three to five cyclesare used to shake down the material's behavior.

a. Since the primary interest from the residual stress analysis is to provideresidual stresses for calculating stress corrosion crack growth undernormal operating conditions, it is desirable to determine a stabilizedresidual stress state that will not change under normal operating cycles.The as-welded residual stresses usually contain localized peak stresses

2 of 15

at some nodal locations. Applying a few operating cycles will stabilize thestress peaks and valleys due to the slight stress redistribution at elevatedtemperatures. It has been determined through experience that theresidual stresses will stabilize after three to five cycles. Five cycles wasused for conservatism.

b. The modeling of the operating cycles involved simultaneously ramping theoperating temperature and pressure between the no load and full loadconditions repeatedly five times. Residual stresses at the fifth operatingcycle were extracted for the crack growth calculation.

c. An example through-wall stress path at the weld/nozzle interface at the90-degree azimuth is used to investigate the residual stresses at eachoperating cycle. The results show that the residual stresses are stabilizedbecause there is virtually no change in the profiles among the five cycles.This is primarily due to the resulting stresses developed in the fabricationprocess.

3 of 15

50

4 0 -

3040 -• Cycle 1

I 0 20 0 Cycle2Pix (" Cycle 3

Cycle4

......................................... ....... ...... Cycle 5

-104

-20 Thickness (inch)


What is the basis for not including an initial weld repair, 50% through-wall?

ENO Response

The presence of an initial weld repair from plant construction (e.g., extending50% of the wall thickness from the inside diameter (ID)) is often assumed whenmodeling Alloy 82/182 piping butt welds. Often for piping butt welds, the residualstress calculated for the ID is a small tensile value, or even compressive, in theabsence of an assumed weld repair. In such cases, the possibility of asignificant weld repair being present on the weld ID can have a relatively largeeffect on the calculated stresses, especially on and near the ID surface.

However, for the Alloy 82/182 branch connection welds at PNP, there are tworeasons why it is not necessary to include a weld repair assumption in theanalysis. First, the design for this weldment specifies a 3600 backweld on the IDsurfaces of the pipe that is about 0.25 inch thick. This design feature results inelevated residual stress levels at the ID surface prior to the post weld heattreatment (PWHT) being applied. The residual stress levels at the inside surfacedue to the presence of the backweld are similar to what would be expected dueto the presence of a weld repair on the ID surface.

Second, any weld repairs would have been made prior to PHWT being applied,and would be expected to extend over a relatively limited circumferential portionof the original weld. Similar to the situation for the elevated residual stressesdue to the presence of the backweld, the PWHT would relax the residualstresses in the weld repair area, including the substantial relaxation expected atthe surface exposed to primary coolant. (The effect of PWHT on the likelihoodof PWSCC initiation occurring on the wetted Alloy 600 and Alloy 82/182 surfaces

4 of 15

of the PNP branch connection weldments was assessed by DominionEngineering, Inc. (see Attachment 2)). Moreover, in the unlikely case thatinitiation occurred in the area of a weld repair, the weld repair would be anadditional source of non-axisymmetric crack loading that would tend to drivecrack growth through-wall over a relatively local circumferential region, ultimatelyresulting in detection of leakage prior to the possibility of unstable pipe rupture.


Provide the repair history on these welds during fabrication.

ENO Response

The manufacturing/quality plan provided in the specification for the primarycoolant system piping provides instructions for performing weld repairs based onthe results of NDE testing. Any defects identified in the nozzle welds would havebeen removed prior to final furnace heat treatment of the assembly. PNP isunable to locate post-fabrication documentation other than the weld radiographstaken after the final furnace heat treatment. These radiographs represent thecondition of the subject welds at the time of installation at the site. A search ofPNP records did not identify any repairs performed on the subject welds sinceinstallation.


Provide detailed geometry information of the weld, including weld thickness.

ENO Response

The weld is 1/2 inch wide at the ID of the hot leg pipe and then increases in widthat a 7- 1/2 0 (+2o/_Oo) angle to the OD of the hot leg pipe. The thickness of the hotleg pipe is 3-3/4 inches.

The details of the nozzle are shown on Palisades Drawing M-1 D-01 08. The hotleg drain nozzle is shown as part number 675-03 as shown below.

5 of 15

.,

0gy rD_)AIN A/Z-i

The weld number in question is designated as 5-675 on the drawing. The detailsshow that it is 1/2 inch wide at the ID of the hotleg pipe and then increases inwidth at a 7- ½2 o (+2o/_0o) angle to the OD of the hot leg pipe.

An exploded view of weld 5-675 follows, showing the additional calculateddimensions, based on a 7.5 degree angle.

6 of 15

3.8 [nYhes

7F 7.5 *2. -0 degrees'0.5 inch,

m

(Opp~en.)

-1-3.76 Inches,

Not t Si 1 6 3F16 inchesd~arncter'

'I-Re¶.- pu .. ui U--: 1 ,-11-0 W s a. RIc ; 'r.sn S

Weld 5-675 width of 3.75 inches was obtained from the hot leg drain nozzle(675-03) installation drawing. The hot leg drain nozzle is located in pipeassembly 673-04 and is attached to pipe number 673-02 as detailed onPalisades drawing M0001 D-0106 and shown below.

7 of 15

PIPE ASSEMBLY 673-04 _.

Pipe Number 673-02 has a specified minimum thickness of 3- 3/ inches asshown in the material table from Palisades drawing MOO01 D-01 06.

- - -- --.

.-t..

.. . . . .

•3-•ri Ti -. - - -"-

-9-

-9-

- Y 9-9-05-Im -5 -Ur4 SAID~qj1 .~ 134F6 EA./D - & - & -£75021 S ~'5SS LiSA ~UISMSAJ? dAEA4PL/AI& AA~2ZU

RJL5 :A.4QZZ -a

~73* LU 73,Zw6.p 4

-673/S-T a~ .1~/S 4g6~ -LO - -jW.43GOWWLI67fSI 6 PW/A-3"6"i -. , CWsA C'

-SS No. AS~48L lM~M 011100- NOMZNALATURMot~ ADTO

LIS Of MATERtIAL - QUANTITIES root ~wI ri>.wjt, ass wAr aN camrAc9

8 of 15


Provide weld bead application sequence.

ENO Response

The weld beads were modeled as full bead rings, with 80 bead rings defined forthe main weld, and seven bead rings defined for the inside diameter patch weldper the labeled sequence shown in the figures below.

CLAD

9 of 15


Provide hardening law chosen in model.

ENO Response

Multilinear isotropic hardening (MISO) modeling was used in ANSYS. Thematerial stress-strain curves were derived using the Ramberg-Osgood curve fitwith the maximum stress modified to be at the flow stress, which is defined asthe average of the yield strength and tensile strength. The approach is astandard practice at Structural Integrity Associates (SI) for weld residual stressfinite element analyses, including the results submitted to the NRC for theInternational Weld Residual Stress Round Robin Phase 2a study, where the SIresults were shown to compare well with the measurements.


Provide weld parameters chosen in model.

ENO Response

The welding parameters used in the ANSYS finite element analyses are asfollows:

Interpass temperatureMelting temperatureAmbient temperatureHeat input both weldsHeat efficiency both weldsInside/Outside heat transfer coefficientInside/Outside temperature

= 350°F= 2500°F= 70OF= 28 kJ/in= 0.8= 5 Btu/hr/ftA2= 70°F


Provide extraction path of WRS profile.

ENO Response

The stress extraction path for the circumferential crack was the radial stresseson the entire weld/nozzle interface area. The stress extraction path for the axialcrack was the hoop stresses within a rectangular 2 inch x 3.75 inch grid locatedon the axial cut plane of the hot leg (i.e., at the 0-degree azimuth).

10 of 15

Radial residual + operating stresses(After PWHT and Hydro Test)

Hoop stress extraction grid for axial crack2" wide x 3.75" tall grid at 0-degree azimuth


Provide weld residual stresses for profiles chosen.

ENO Response

Requested stress outputs are included in the supporting fileRAISupportFiles.zip, which are provided in the Enclosure.

11 of 15

Radial stresses (SX) for the circumferential crack analysis were extracted at theelement face centroid location for each element on the crack face. The outputswere saved to two files:

Crack6_COORDl.txt contains the element face centroid location.

STR_FieldOper_61.txt contains the stresses for each element on thecrack face, where the radial component (SX) was used in thecircumferential crack analysis.

Hoop stresses (S-hoop) for the axial crack analysis were output in a tabularformat in terms of Radius, Height, S-hoop, in the attached computer fileHoop_0.csv.


Was the WRS profile fit with a polynomial? If so, what order was thepolynomial? If not, how were K values derived?

ENO Response

For the circumferential crack, the stress intensity factors, including the residualstresses, were calculated via deterministic finite element analyses using linearelastic fracture mechanics (LEFM) principles. Seven fracture mechanics finiteelement models are derived to include "collapsed" crack meshing that representfull circumferential flaws surrounding the nozzle at various depths within the bossweld. The depths vary from 0.13 inch to 3.95 inch.

High order SOLID95 elements in ANSYS were used around the crack tips, withthe mid-side nodes shifted to the "quarter point" locations to properly capture thesingularities at the crack tips. The 3.95 inches deep crack is shown as anexample.

In order to determine the "Ks," the radial component (relative to the nozzle)residual stresses, plus operating stresses at the boss weld/nozzle interface, wereextracted from the residual stress analysis and reapplied on the crack face assurface pressure loading. The transferred radial stresses onto the crack face arealso shown in the figure below.

12 of 15

The crack face pressure transfer approach is based on the load superpositionprinciple: Anderson, T. L., "Fracture Mechanics Fundamentals and Applications,"Second Edition, CRC Press, 1995. The superposition technique is based on theprinciple that, in the linear elastic regime, stress intensity factors of the samemode, which are due to different loads, are additive (similar to stresscomponents in the same direction). The superposition method can besummarized with the following sketches:

13 of 15

P(x)

(a)P(x) P

: . ' ' p(x) ......

(b) (c) (d)

A load p(x) on an uncracked body, Sketch (a), produces a normal stressdistribution p(x) on Plane A-B. The superposition principle is illustrated bySketches (b), (c), and (d) of the same body with a crack at Plane A-B. Thestress intensity factors resulting from these loading cases are such that:

KI(b) = KI(c) + KI(d)

Thus, KI(d) = 0 because the crack is closed, and:

KI(b) = KI(c)

This means that the stress intensity factor obtained from subjecting the crackedbody to a nominal load p(x) is equal to the stress intensity factor resulting fromloading the crack faces with the same stress distribution p(x) at the cracklocation in the uncracked body.

For the axial crack, all the stresses (including residual) were calculated usingfinite element analyses. These stresses were then used to calculate stressintensity factors using a buried elliptical crack model by use of influencefunctions. Fitting stresses to polynomials was not required. A published Ksolution was used: P. M. Besuner, "Residual Life Estimates for Structures withPartial Thickness Cracks," ASTM STP 590, 1974. This K solution is available intwo commercial software packages - NASCRAC and SmartCrack. SmartCrackwas used for the analysis herein.

14 of 15


Provide the Structural Integrity Associates report for the flaw evaluation for thesenozzle welds.

ENO Response

The Structural Integrity Associates, Inc. calculations will be provided in asupplemental RAI response letter.

15 of 15

ATTACHMENT 2

Dominion Engineering, Inc.Letter No. L-4199-00-01

Effect of Post-Weld Heat Treatment Applied to Alloy 82/182 Full-PenetrationBranch Pipe Connection Welds at Palisades

February 25, 2014

5 Pages Follow

Dominion [n~ineerin , Incw

February 25, 2014L-4199-00-01, Rev. 0

Mr. William SimsEntergy Operations, Inc.1340 Echelon ParkwayJackson, MS 39213

Subject: Effect of Post-Weld Heat Treatment Applied to Alloy 82/182 Full-Penetration

Branch Pipe Connection Welds at Palisades

Dear Mr. Sims:

Introduction

At Palisades, Alloy 82/182 full-penetration dissimilar metal V-welds were used for installationof Alloy 600 branch connection nozzles on the hot-leg and cold-leg piping. Subsequent toinstallation of these nozzles, a post-weld heat treatment (PWHT) was performed of the adjacentcarbon steel material. Because the PWHT was performed subsequent to installation of thesenozzles, the Alloy 82/182 and Alloy 600 materials also were exposed to this heat treatment.There is one such nozzle on the hot-leg piping at Palisades (2-inch Hot Leg Drain Nozzle) andeight such nozzles on the cold-leg piping at Palisades (six 2-inch nozzles and two 3-inchnozzles). Direct visual examinations of all these weldments for evidence of leakage have beenpreviously performed per the requirements of ASME Code Case N-722-1 [I] for each location,with no such evidence reported.

The purpose of this letter is to assess the effect of the PWHT on the susceptibility of the nickel-base Alloy 82/182 weld metal and Alloy 600 base metal to primary water stress corrosioncracking (PWSCC).

Benefit of Post-Weld Heat Treatment (PWHT) for Resistance to PWSCC Initiation

Testing and extensive plant experience demonstrate that Alloy 600 weldments that have beenexposed to PWHT after welding have greatly reduced susceptibility to occurrence of PWSCC:

" In 2007, Scott ([2] and [3]) reported that PWHT of test samples of Alloy 182 welds wasfound to greatly delay occurrence of PWSCC. This work included detailed laboratoryinvestigation of the effect of the PWHT on residual stresses and the materialmicrostructure of the nickel-base weld, including stress measurements on mockups.Scott [2] concluded that the large benefit of PWHT is due to the reduction in residualstress levels it produces, including very significant relaxation of the residual stress in thesurface layer of the weldment, and that the reason for "the strong favorable effect onsurface residual stresses and consequently on PWSCC initiation resistance" is that thePWHT causes recrystallization of the surface cold-worked layer produced by grinding.

* Consistent with the laboratory work, PWR plant experience shows hundreds of cases ofPWSCC when the Alloy 82/182 or Alloy 600 material was not exposed to PWHT, butextremely few cases when the material was exposed to PWHT subsequent to welding:

12100 Sunrise Valley Drive, Suite 220 1 Reston, VA 20191 U PH 703.657.7300 0 FX 703.657.7301

Dominion [nineerin?, Inc. L-4199-00-01, Rev. 0p. 2 of 5

- There have been more than 120 cases of PWSCC detected in small-diameter Alloy600 instrumentation nozzles and pressurizer heater sleeves in plants designed byCombustion Engineering ([4], [5], and [6]). These nozzles in CE-designed plantswere not exposed to PWHT after installation. On the other hand, small-diameterAlloy 600 instrumentation nozzles are also used extensively in B&W-designedPWRs, and there have been extremely few cases of PWSCC reported for thesecomponents, which were installed prior to PWHT being performed [4]. The limitedcases of PWSCC reported for a component exposed to PWHT include one reactorvessel bottom-mounted nozzle at Gravelines Unit 1 in France [7]. In 2011, PWSCCwas reported for one bottom-mounted nozzle at this plant associated with pre-existingbase metal manufacturing defects.

- In a 2007 paper., Scott, et al., [2] cited the very favorable plant experience that therehad been no occurrences of PWSCC in Alloy 600 weldments in France where theweldment had been subjected to a PWHT.

- Bartsch, et al., reported in 2003 that no PWSCC had been detected in the CRDMnozzles at Obrigheim, despite their having been made of Alloy 600 and being weldedwith Alloy 82/182 J-groove welds similar to those that had cracked at plants in othercountries and for which PWHT was not applied subsequent to nozzle installation [8].The freedom from cracking was attributed to the reactor vessel head having beenPWHT after installation of the CRDM nozzles.

- Papers by Bibollet, et al., [9] and Verdiere, et al. [10] in 2006 describe the occurrenceof PWSCC in a few steam generator divider plates in France. A main observationregarding this PWSCC is that it has essentially all been associated with the dividerplate to stub weld, which is the only weld in the divider plate assembly that was notgiven a PWHT. An exception is for one case where PWSCC occurred as a result ofcold work due to impacts by a large loose part.

Based on the favorable test data and plant experience for Alloy 600/82/182 components exposedto PWHT, it is unlikely that PWSCC initiation of a flaw of engineering size (i.e. depth of I or 2millimeters) has occurred on the Alloy 82/182 full-penetration branch pipe connection welds atPalisades. Eight of these nine locations operate at reactor cold-leg temperature, and the other atreactor hot-leg temperature. None of them operates at pressurizer temperature. Because of therelatively large effect of operating temperature on PWSCC susceptibility (e.g., [11]), theoperating temperatures applicable to these components add confidence to the conclusion thatPWSCC initiation is unlikely to have occurred.

In the case of the single nozzle operating at reactor hot-leg temperature, detailed weld residualstress calculations have been performed simulating the effect of welding and the other sources ofstress applicable to normal operating conditions. The finite-element simulation also considersthe effect of the PWHT applied subsequent to installation of the nozzle. We understand thatthese results show a maximum tensile radial stress over the wetted surface of the nickel-basematerials of 30.4 ksi, and a maximum tensile hoop stress over the wetted surface of the nickel-base materials of 24.3 ksi.

These calculated stresses are relatively modest compared to the surface stresses of 40 to 60 ksithat are typically predicted for components that have been reported to experience PWSCC.Laboratory work shows that the time to PWSCC initiation is sensitive to the surface stress level,

Dominion [nyineerin?, In(. L-4199-00-01, Rev. 0p. 3 of 5

with the sensitivity typically described using an exponent on the stress in the range of 4 to 7 [12].Furthermore, while it is considered that there is no firm "threshold" below which PWSCC willnever occur, from a practical experience perspective a tensile stress of about 20 ksi (138 MPa) isconsidered a conservative lower bound estimate of a stress level below which PWSCC initiationwill not occur during plant lifetimes [13]. (The 20 ksi threshold stress corresponds to about 80%of the lower bound yield strength for Alloy 600 materials at operating temperatures.)

Thus, the weld residual stress results specific to the single nozzle operating at reactor hot-legtemperature further supports plant experience and test data. Given that the calculated peaksurface stress is reasonably close to the expected threshold for initiation during plant timescalesand given the large demonstrated sensitivity of initiation time to surface stress, there is a lowprobability that PWSCC initiation of a flaw of engineering size has occurred on the single Alloy82/182 full-penetration branch pipe connection weld at Palisades. Because of the cold-legoperating temperature of the nozzles located on the Palisades cold legs and because the weldresidual stress is expected to be similar for the cold-leg locations, this conclusion extends to theeight cold-leg nozzles.

Benefit of PWHT for Resistance to PWSCC Growth Rate

In the unlikely case that crack initiation has occurred, the effect of PWHT on the resistance toPWSCC crack growth would become a factor. Section 3.1.8 of MRP-l 15 [14] discusses theeffect of PWHT on the PWSCC crack growth rate. Cassagne et al. [15] reported that there was abeneficial effect of PWHT (by a factor between two and four) on CGRs measured for four weldmetals with various carbon and silicon contents. Based primarily on these data, Le Hong et al.[16] recommend that crack growth rates be reduced by a factor of 2.0 for stress relievedspecimens compared to otherwise similar as-received specimens.

The standard deterministic crack growth rate equation for Alloy 182 weld metal in MRP-1 15[14] was developed on the basis of specimens that were not exposed to PWHT. Thus,calculations performed using the deterministic MRP-1 15 equation tends to result inconservatively high crack growth rates when applied to welds that were exposed to a PWHT.This level of conservatism can be considered in terms of the statistical variability in crack growthrates exhibited in MRP-l 15 across different test welds for equivalent temperature and stressconditions. The deterministic equation for Alloy 182 corresponds to the 7 5 th percentile of thelog-normal distribution describing this material variability. The 9 5 th percentile of thisdistribution corresponds to a crack growth rate that is 1.77 times the 7 5 th percentile value. Thus,in effect the deterministic crack growth rate equation for Alloy 182 corresponds to a crackgrowth rate above the 951h percentile of behavior if the effect of PWHT is credited.

Conclusions

In summary, there is a low probability that a PWSCC crack of engineering size has initiated onthe Alloy 82/182 full-penetration branch pipe connection welds at Palisades. Confidence in thisconclusion is provided by a combination of laboratory investigations, extensive plant experience,the favorable operating temperatures, and finite-element weld residual stress calculations specificto Palisades.

In the unlikely case that crack initiation were to occur, there is expected to be a significantbeneficial effect of the PWHT on the crack growth rate. Furthermore, in the unlikely event thatcircumferential PWSCC were to occur in the Alloy 82/182 full-penetration weld, the significant

Dominion [n?ineerin?, Inc. L-4199-00-01, Rev. 0p. 4 of 5

variability in residual stress with azimuthal position around the nozzle (for the geometry of anozzle welded into a cylindrical pipe) would tend to drive crack growth through-wall along partof the circumference, tending to support detection of leakage prior to the possibility of unstablepipe rupture. The periodic visual examinations for evidence of leakage that are performed foreach location per ASME Code Case N-722-l [I], including during every refueling outage for thesingle hot-leg nozzle, are direct examinations of the metal surface that are capable of detectingsmall amounts of pressure boundary leakage. In the unlikely event that axial PWSCC were tooccur in the Alloy 82/182 full-penetration weld and the adjoining Alloy 600 nozzle, thestructural stability provided by the pipe branch connection geometry would also tend to supportdetection of leakage prior to the possibility of unstable pipe rupture.

Finally, the potential presence of weld repairs made during plant construction would not affectthese conclusions. Any such weld repairs would have been made prior to PHWT being applied,and would be expected to extend over a relatively limited circumferential portion of the originalweld. The PWHT would relax the residual stresses in the weld repair area, including thesubstantial relaxation expected at the surface exposed to primary coolant. Moreover, in theunlikely case that initiation occurred in the area of a weld repair, the weld repair would be anadditional source of nonaxisymmetric crack loading that would tend to drive crack growththrough-wall over a relatively local circumferential region, again tending to support detection ofleakage prior to the possibility of unstable pipe rupture.

If you have any questions regarding this topic, please do not hesitate to contact me at (703) 657-7315 or gwhiteadomeng.com, or Mr. John Broussard at (703) 657-7316 oribroussard&Tdomeng.com.

Sincerely,

Glenn A. White, P.E.Principal Engineer

References

1. ASME Code Case N-722-1, "Additional Examinations for PWR Pressure Retaining Welds inClass 1 Components Fabricated With Alloy 600/82/182 Materials, Section XI, Division I"ASME, January 26, 2009.

2. P. Scott, et al., "Comparison of Laboratory and Field Experience of PWSCC in Alloy 182Weld Metal," Proceedings of the 13' International Conference on EnvironmentalDegradation of Materials in Nuclear Power Systems, paper 25, CNS, 2007.

3. 0. Calonne, M. Foucault, P. Combrade, and P. Scott, "Stress Corrosion Cracking of Alloy182 Ni-Base Alloy Weld Metal," Program on Technology Innovation: Proceedings-2007AECL/COGiEPRI Workshop on Cold Work in Iron- and Nickel-Base Alloys Exposed to HighTemperature Water Environments, June 3-8, 2007, Mississauga, Ontario, EPRI, Palo Alto,CA: 2008. 1016519.

4. Materials Reliability Program: PWSCC of Alloy 600 Type Materials in Non-SteamGenerator Tubing Applications -- Survey Report Through June 2002: Part 1: PWSCC in

Dominion [nineerin?, Inc L-4199-00-01, Rev. 0p. 5of5

Components Other than CRDM/CEDM Penetrations (MRP-87), EPRI, Palo Alto, CA: 2003.1007832. [freely available for download at www.epri.com]

5. W. Bamford and J. Hall, "A Review of Alloy 600 Cracking in Operating Nuclear Plants:Historical Experience and Future Trends," Proceedings of the 11Ph International Conferenceon Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors,p1071-1081, ANS, 2003.

6. W. Bamford and J. Hall, "A Review of Alloy 600 Cracking in Operating Nuclear PlantsIncluding Alloy 82 and 182 Weld Behavior," Proceedings of ICONEI2 12th InternationalConference on Nuclear Engineering, ICONE12-49520, ASME, 2004.

7. "Compte Rendu de la Rdunion de la Commission 'Technique' de la CLI de Gravelines du 07f~vrier 2012," available at http://www.cli-gravelines.fr/content/download/3151/40041 /version/I /file/CLITech_2012-02-07_CR.pdf.

8. R. Bartsch, et al., "Integrity of RPV Head Penetration Nozzles in the German NPPObrigheim," Proceedings of the I11h International Conference on EnvironmentalDegradation of Materials in Nuclear Power Systems-Water Reactors, p 1082-1089, ANS,2003.

9. C. Bibollet, et al., "EDF Field Experience on Steam Generator Divider Plate Examinations,"Proceedings of the International Symposium Fontevraud 6 on Contribution of MaterialsInvestigations to Improve the Safety and Performance of L WRs, SFEN, paper Al 38-TO4a,September 2006.

10. N. Verdi~re, et al., "Maintenance Strategy of Inconel Components in French PWRs,"Proceedings of the International Symposium Fontevraud 6 on Contribution of MaterialsInvestigations to Improve the Safety and Performance of L WRs, SFEN, paper Al 26-TO4a,September 2006.

11. Materials Reliability Program: Primary System Piping Butt Weld Inspection and EvaluationGuideline (MRP-139, Revision 1), EPRI, Palo Alto, CA: 2008. 1015009. [freely available fordownload at www.epri.com]

12. PWSCC Prediction Guidelines, EPRI, Palo Alto, CA: 1994. TR-104030. [freely available fordownload at www.epri.com]

13. Materials Reliability Program. Generic Evaluation of Examination Coverage Requirementsfor Reactor Pressure Vessel Head Penetration Nozzles, Revision I (MRP-95R1), EPRI, PaloAlto, CA: 2004. 1011225.

14. Materials Reliability Program. Crack Growth Rates for Evaluating Primary Water StressCorrosion Cracking (PWSCC) ofAlloy 82, 182, and 132 Welds (MRP-115), EPRI, Palo Alto,CA: 2004. 1006696. [freely available for download at www.epri.com]

15. T. Cassagne, D. Caron, J. Daret, and Y. Lef~vre, "Stress Corrosion Crack Growth RateMeasurements in Alloys 600 and 182 in Primary Water Loops under Constant Load,"Proceedings of 9,h International Symposium on Environmental Degradation of Materials inNuclear Systems-Water Reactors, The Minerals, Metals & Materials Society, Warrendale,PA, 1999, pp. 217-224.

16. S. Le Hong, J. M. Boursier, C. Amzallag, and J. Daret, "Measurements of Stress CorrosionCracking Growth Rates in Weld Alloy 182 in Primary Water of PWR," Proceedings ofl 0"'h

International Conference on Environmental Degradation of Materials in Nuclear PowerSystems--Water Reactors, NACE International, 2002.

ENCLOSURE

ENO Response to Question RAI-1.11

The files discussed in the ENO response to RAI-1.11 (repeated below) in Attachment 1are provided in this Enclosure.


Provide weld residual stresses for profiles chosen.

ENO Response

Requested stress outputs are included in the supporting file AlSupportFiles.zipand are transmitted via the enclosure.

Radial stresses (SX) for the circumferential crack analysis were extracted at theelement face centroid location for each element on the crack face. The outputswere saved to two files:

Crack6_COORD1.txt contains the element face centroid location.

STRFieldOper_61.txt contains the stresses for each element on thecrack face, where the radial component (SX) was used in thecircumferential crack analysis.

Hoop stresses (S-hoop) for the axial crack analysis were output in a tabularformat in terms of Radius, Height, S-hoop in the attached computer fileHoopO.csv.

(Electronic Files Enclosed)

otto w. gustafson licensing manager · (weld number 5-675) consisted of an intermediate post weld...

Documents